umairnadeem123

10 ms·

1.
by umairnadeem123 7mo ago
yeah you can see exactly why a comment/post scores the way it does. case is one of the factors
2.
by umairnadeem123 7mo ago
IMO this is solving the wrong problem. the session log is just noise - its like attaching your google search history to a stackoverflow answer to "prove" you did the research. nobody wants to read 500 lines of an agent going back
3.
Show HN: HN Bot Detector - Detects LLM-Generated Comments on Hacker News(github.com)
7 points by umairnadeem123 7mo ago|3 comments
4.
by umairnadeem123 7mo ago
almost definitely AI-generated: https://hn-bot-detector.vercel.app/s/e7Qhw8yh
5.
by umairnadeem123 7mo ago
> I tried to avoid writing this for a long time, but I'm convinced MCP provides no real-world benefit IMO this is 100% correct and I'm glad someone finally said it. I run AI agents that control my entire dev workflow through sh
6.
by umairnadeem123 7mo ago
NGL the flaw in this piece is the same flaw in every "AI will replace X" argument - it assumes the bottleneck was ever the typing.it wasn't. the bottleneck is knowing what to build and why.I use AI agents for probably 80% of
7.
The Long Afterlife of the Console Modchip(hackaday.com)
2 points by umairnadeem123 7mo ago|0 comments
8.
by umairnadeem123 7mo ago
yikes- fair points on being skeptical of arxiv quality, thats a real problem. IMO the content of the paper either holds up or it doesnt regardless of who wrote it
9.
by umairnadeem123 7mo ago
practical takeaway for builders: agent context docs seem to help only when they contain non-obvious project constraints, not generic rules. in my own workflows, the useful bits are things like failure patterns/edge cases or antipattern
10.
by umairnadeem123 7mo ago
? not sure i follow
11.
Codified Context for AI Agents in a Complex Codebase(arxiv.org)
3 points by umairnadeem123 7mo ago|6 comments
12.
Shifting Security Left for AI Agents with GitGuardian MCP(blog.gitguardian.com)
2 points by umairnadeem123 7mo ago|0 comments
13.
by umairnadeem123 7mo ago
def useful to show what models recommend in real use (over just meaningless benchmarks), but i still think small prompt wording and repo setup changes can change the outcome quite a bit so id love tighter controls there. having tried claude
14.
by umairnadeem123 7mo ago
this is what happens when a "public" key type quietly turns into a privileged key type without forcing people to re-scope it, not really a dev mistake IMO, it's a platform design bug and google needs hard separation between p
15.
Security Boundaries in Agentic Architectures(vercel.com)
1 points by umairnadeem123 7mo ago|0 comments
16.
by umairnadeem123 7mo ago
honestly the best way to beat my detector is to just write worse
17.
by umairnadeem123 7mo ago
ah yes. looks like it was going over the max token limit. if you retry, will work again!
18.
by umairnadeem123 7mo ago
I work at a large social media company - it has become a popularity contest on who can ship more AI slop the fastest. Real productivity gains are questionable tbh. Slop is replacing more slop.
19.
by umairnadeem123 7mo ago
something deeply ironic about this.
20.
Show HN: I built an LLM comment detector for HN (I got banned)
4 points by umairnadeem123 7mo ago|6 comments
21.
by umairnadeem123 8mo ago
the connection draining feature is really clever. weve dealt with connection storms from app crashes before and the standard playbook is just to restart everything and hope the thundering herd doesnt immediately repeat. having the proxy abs